{T}
TokenMath.net
Updated 2026-09-22 11:16 UTC
Head-to-Head Specification Battle

Llama 4 Scout (17B Active / 109B MoE) — Together AI vs Codestral (Current Generation)

Side-by-side technical and economic comparison between Meta / Together's Llama 4 Scout (17B Active / 109B MoE) — Together AI and Mistral AI's Codestral (Current Generation).

Endpoint Availability & Transition Notice

Llama 4 Scout (17B Active / 109B MoE) — Together AI: Classified as a Legacy Reference model (Deprecated on Together AI (Reference)).

Core Technical & Pricing Comparison
MetricLlama 4 Scout (17B Active / 109B MoE) — Together AICodestral (Current Generation)Advantage
Provider OrganizationMeta / TogetherMistral AI
Standard Input / 1M$0.18$0.30Llama 4 Scout (17B Active / 109B MoE) — Together AI (40% lower)
Cached Input / 1MNone$0.03Codestral (Current Generation) lower
Output / 1M$0.59$0.90Llama 4 Scout (17B Active / 109B MoE) — Together AI lower
Context Window328k256kLlama 4 Scout (17B Active / 109B MoE) — Together AI (328k)
Max Generation Tokens16.4k8.2kLlama 4 Scout (17B Active / 109B MoE) — Together AI
Tokenizer FamilyMeta Llama 3/4 Tiktoken (128k vocabulary)Mistral Tekken BPE (~131k vocabulary)

Interactive Side-by-Side Cost Simulator

Live Delta
Input Tokens / Req2,500
Output Tokens / Req800
Daily Requests10,000
Llama 4 Scout (17B Active / 109B MoE) — Together AILOWER COST
$276.60 / month
$0.00092 / request
Codestral (Current Generation)
$339.75 / month
$0.00113 / request
Selecting Llama 4 Scout (17B Active / 109B MoE) — Together AI saves $63.15 / month (18.6% savings) over Codestral (Current Generation).

When to Choose Llama 4 Scout (17B Active / 109B MoE) — Together AI

Opt for Llama 4 Scout (17B Active / 109B MoE) — Together AI when your engineering requirements prioritize Meta / Together's ecosystem, specific tokenizer efficiencies (Calibrated Llama 3 Tokenizer (±3%)), or when your expected prompt-to-completion ratios favor its $0.18/M input rate.

When to Choose Codestral (Current Generation)

Opt for Codestral (Current Generation) when looking for Mistral AI's tooling integration, specific context window depth (256k tokens), or when output generation volume favors its $0.9/M rate.